C64 Reloaded MK2, Ultimate 1541 and a 1541-II

Caution: Non registered users only see threads and messages in the currently selected language, which is determined by their browser. Please create an account and log in to see all content by default. This is a limitation of the forum software.


Also users that are not logged in can not create new threads. This is a, unfortunately needed, counter measure against spam. Please create an account and log in to start new threads.

Don't Panic. Please wash hands.
  • Hi all, just a little problem I have that I'll try to explain as best I can. I have also posted this on Gideon's group on Facebook,

    I have a C64 Reloaded MK2, 1541 Ultimate 2 and a 1541-II drive. When I have them all hooked up on the IEC chain (1541-II off) the Ultimate can't load anything from a mounted image. It can do a mount+run okay though. This works fine on a real C64. If I have the 1541-II on and mount a image on device 9 using the Ultimate, it will load fine. It seems there is a problem when the 1541 real drives are turned off but present in the IEC chain. Anyone else experienced this?

  • It seems there is a problem when the 1541 real drives are turned off but present in the IEC chain.

    Right, and that's expected: Every IEC device has 1k pull-up resistos on the IEC signals, where "up" is always the local 5V rail. If the drive is turned off, these resistors become pull-down resistors, as the 5V rail of the switched-off device is now GND. This is known to cause trouble since 1981. Just because it works in one configuration doesn't mean it works with a different computer. A classic example of "how to build a logical-perhaps-gate" :-)

  • Quote

    It seems there is a problem when the 1541 real drives are turned off but present in the IEC chain.

    Yes there is, you shouldn't be doing this, the IEC bus is not designed for this. All devices connected to the bus must be turned on!

  • Okay, I just had to do a test as I have plenty of time to do so these days. I pulled three C64's out of storage, one breadbin and two C64c's. They all worked perfectly with the drive turned off in the middle of the IEC chain. No big deal as I love the Reloaded MK2, and use it a lot, but interesting nevertheless.


  • I pulled three C64's out of storage, one breadbin and two C64c's. They all worked perfectly with the drive turned off in the middle of the IEC chain.

    Do you have a possibility to pull the CIA chip on U2 out of one of these "working-by-accident" units and put them into the MK2?


    And can you measure the voltage inside the computer to see if there's a difference to what the MK2 generates on-board?


    With a single IEC device connected, but not switched on, this may be an "on the edge" case, and both the CIA chips' date code and the supply voltage may have an influence (lower C64 voltage may make it work). Not that it really matters - it's not a recommended way of operation for almost 40 years, but if you want to dedicate your time to an approach of slightly more scientific nature, the C64-community will surely appreciate new knowledge.


    The biggest difference between the old C64 boards and the C64 Reloaded MK2 is that the three lines ATN, DAT and CLK have dual open-collector drivers on the C64RMK2, where the original boards only have a single one. This ensures proper operation with a high amount of IEC devices. Rememebr that it's perfectly possible to connect four floppy drives (units 8-11), a printer (unit 4) and a plotter (unit 5, total of six IEC units), where the original boards start to fail. The MK2 doesn't, because it has twice the driver power when pulling signals to GND.


    This should not make a difference in "listen" mode, as the drivers are high-impedance in that state. However, the slightly-different leakage current might tip the scale in your case. Worth investigating if you ask me!


    Another difference is the clamping diodes that I have added for further CIA protection. These are standard BAV99, and they do pretty much nothing when the IEC bus is in it's normal operational state. They only conduct if a voltage is lower than GND or higher than VCC. On normal operation, they add a tiny bit of capacitance, but that should be the same order of magnitude that a cable will add, so "nothing to worry" either. I definitely prefer the protection property over compatibility with an unrecommended edge case.


    Another thing worth asking is the exact resistor value that Gideon has chosen as a pull-up. The standard value that was used back then was 1k, and if Gideon has chosen somehing different, it might explain different behaviour. Same goes for the input stage on the 1541U: If it's a standard 3.3V CPLD, the switching threshold will be 1.65V, where the threshold for the CIA chips is documented to be 1.4V. Again, with the 1541U powered by the C64RMK2, a slightly different voltage might explain different behaviour as well - I'd expect the 5.0V of the C64RMK2 to be higher than many ageing C64 PSUs that even tend to drop voltage on higher load. While the 3.3V rail of the 1541U won't change, the pull-up current to 5V will change with supply voltage.


    So.. a complex topic, but well worth investigating!

  • Definitely interested in investigating this further. I know that one of my C64's has every chip socketed so pulling chips will be easy. It might be worth mentioning that the chips that now live in my Reloaded are from another C64c that used to work until the SuperPLA (shortboard) decided to clock out for good. This machine also worked with the drive in the chain while off. I'll get some readings and report back, probably over the weekend as I have two days of work.. If you could give me a comprehensive list of tests and methods I can carry out so I can be as accurate as possible with the results would be great. My email address is linked to order 83691 if that makes it easier to correspond. I am better with the practical than the theory, but I'm always keen to learn. and add knowledge to our community.

  • Okay, some results:


    Reloaded MKII CIA - 4.89V (Indivision Power Supply)

    Commodore 64 CIA - 4.71V (Original Power Brick with inline Power Saver)


    I did the CIA chip swap, same issue was present as before.

    As I have stated before, the chips in the Reloaded are from a C64 that worked with the configuration in question.

  • The lower voltage may already be the explanation - it shifts both the input threshold and the pull-up current.


    Can you explain what an "inline power saver" is? Is that something on the secondary side?

  • Oh - and "HQ" stands for "highly questionable" ? How does it make sure that there's no voltage drop in the circuit? I'd have chosen the good old Z-diode approach. Granted, you wouldn't have the LEDs, but that product is pretty useless, as I don't see any indication of an insurance for a possible failure. You've paid for safety, but you got nobody to talk to if something fails.


    How does it behave if you remove the in-line thingy? Is the 5V rail coming a bit higher then? I mean, you're beyond the recommended 5% deviation, so "anything" can happen.

  • Errr righto, firstly you didn't know what it was and now you are an expert on it without having one to test in the lab to see if it does actually work. Also it's about protecting from overvoltage not under voltage, like a circuit breaker to protect the C64. Did you actually read the description? I just put that link you help give you an idea of what it was, not the exact one I have but similar in operation. Don't get cranky at me when I'm trying to help just because your clone C64 doesn't operate the same as a real one. Did you have a real C64 on the test bench to reverse engineer or just take a guess and throw a few good old Z-diodes in the Reloaded to make it work?

  • Quote

    Don't get cranky at me when I'm trying to help just because your clone C64 doesn't operate the same as a real one.

    Non powered devices cause random problems with all C64s - it's not a problem of the reloaded.


    Regardless, please do what jens ask for - remove that device and try again. 4,7V really is too low, you should get something close to 4,9V - like with the reloaded,

  • Hmmm strange that I have had four C64's of different design work the same way with or without the device inline. I will get a reading tomorrow, but I suspect they will be the same result.

  • I think most people back in the day had one disk drive and maybe a printer, which would have been off but still connected. Anyway I have some more measurements to ponder. This time I also included a different power supply as well.


    White brick type Power Supply full or resin (same one as before)

    Output from Power Supply unloaded - 5.2V

    CIA - 4.74V

    CIA with C64 Saver - 4.71V


    Black type Power Supply old transformer style, no resin

    Output from Power Supply unloaded - 5.2V

    CIA - 4.92V

    CIA with C64 Saver - 4.85V


    In all scenarios the C64 worked without issue with the drive turn off connected on the IEC chain.

    Also voltage readings taken didn't change at all whether the disk drive was on or not.

  • Errr righto, firstly you didn't know what it was and now you are an expert on it without having one to test in the lab to see if it does actually work.

    I know the circuit behind it - a very simple one. And the product is *very* close to snake oil if it does not provide additional protection, such as financial help in case of a failure. Ask the supplier if he has a product liability insurance for it - my bet is on "no", with the sincere intention of apologizing if there's a different answer.


    I think most people back in the day had one disk drive and maybe a printer, which would have been off but still connected.

    If you're taking that as a comparison verbally, then please use it as a comparison for your trials as well. To my knowledge, you're still using the 1541U, which is a variable that we haven't excluded yet. The 1541U is a CMOS device, whereas the originals are NMOS/HMOS devices, behaving differently when switched off, and having different switching thresholds. We've already establiched that the threshold difference is 0,25, so we might as well stop investigating here.


    Further, printers have 74xx TTL chips as their IEC stages, which behave different from the VIA chips of the 1541.


    Once again, I'm all for finding the actual difference, but we need to apply scientific standards when investigating this. As Tobias pointed out correctly: Errors with a switched-off device on the IEC bus are pretty random and expected. They are mentioned in the original 1541 manual, so the effect you're observing with your C64 Reloaded MK2 is within the original specifications.

  • Lmao, really? I think you'll find that more than half the products on the market at the moment aren't covered by any sort of liability insurance as the work is done by retro enthusiasts for other retro enthusiasts. If the circuit is so simple, given that you previously never heard of such a device, then humour me and tell me how 'your' version would work.


    I don't own a Commodore MPS 1200 printer anymore so I can't conducts any tests on that for you unfortunately. I smashed it back in 1992 when the blasted thing wouldn't print my essay that was due the next day for school lol.


    Look Jens, I feel that because I have pointed out an operational difference between the Reloaded MKII and a real C64 you have taken it personally. This was never meant to be the case. If I thought the whole Reloaded project was shit then I would have never bought one in the first place. It is a fantastic product and I use it as my main C64 machine. I was just pointing out an observation that I noticed, and did say it's no big deal. But the point does still remain - 4 different C64's, two different power supplies, and the issue is NOT there. Maybe you could buy/borrow/steal a 1541 Ultimate 2 (I have the non plus version) and do further testing yourself.


    I honestly feel if anyone is under attack, it's me. I have tried to help out and have been a good customer of iComp products (because they are good products and we, the enthusiasts, value them). You no doubt can see what I have bought and the list is almost as long again with iComp items purchased through AmigaKit. Good way to alienate customers with such an attitude. Anyway, if there is anyway I still can further assist, I will!

  • If the circuit is so simple, given that you previously never heard of such a device, then humour me and tell me how 'your' version would work.

    I didn't hear of the name that the person gave it.


    I did hear of a crowbar circuit before. It's basic electrical engineering, and it's over-engineered for this application. For a simple 5V protection, a 3W Zener diode with 5.1V zener voltage will do. These tend to melt and short out if overloaded, which is a known-good protection against over-voltage and wrong polarity. Costs about 20 cents if bought in single pieces, maybe 3 cents if bought on a spool.


    I think you'll find that more than half the products on the market at the moment aren't covered by any sort of liability insurance

    Thanks for pointing out the difference between hobbyist-offerings on eBay that mostly don't even come from registered businesses, and a professional company. Especially if you're offering "safety" (as that product indicates), it's a shame that there's practically no warranty with it. Granted, the main reason that iComp has product liability insurance is that the insurance usually fights off claims. However, if we really mess up at some point, I believe that customers should not be left with nothing if iComp closes doors forever because of a valid claim. With most of the gear we deal with in this hobby getting harder and harder (=more expensive) to replace, the importance of such an insurance should be easy to understand.


    I honestly feel if anyone is under attack, it's me.

    Apologies, that wasn't my intention. I can put such a test on Tobias' todo-list, but he's currently on a different project. Trouble is that Tobias is working remotely, so he doesn't have access to my measurement gear, which will be required to find the subtle difference that we're most likely looking for. With more data from your side, such as testing with a printer, plotter or just a second "real" 1541, it would have shortened the test-time for us.

  • Actually I can help you there. I have another two working 1541 disk drives in my storage if that will help with tests. I just hope I have enough cables.

    How would you like me to set them up to test to get the results your after? Put them all in the chain while being off to see what happens?

    My place will look like a copy party from the 80's lol ;)

  • The last reply was more than 365 days ago, this thread is most likely obsolete. It is recommended to create a new thread instead.